Well we are not going to detail out how actually this works but we have got information regarding the similar patent which Google filed in September of last year but published only a few days ago.
As Google X driverless cars are running since 2009 and you can also check manuals actually how things in it had worked until now. But just a few days ago we have got new patent information filed by Google date back in September 2013 which allows driverless cars to make it possible to scan and read traffic signals.
In order to increase its accuracy Google has added various methods to determine the traffic lights. Here’s how Phandroid describes, “As detailed in the patent filing, the software works by first receiving the images taken by the camera, filtering images (those taken of intersections), classifying images by searching for red, yellow, and green blobs where it can then label the classified colored blobs accordingly. Using motion compensation to identify associated labels, it can determine the 3D location of traffic signals and generate traffic signals on a map.”
So guys it seems to be like Google is also shooting balls for future for better accuracy, performance and integration with its services.